Meta-analysis for surrogacy: accelerated failure time models and semicompeting risks modeling

There has been great recent interest in the medical and statistical literature in the assessment and validation of surrogate endpoints as proxies for clinical endpoints in medical studies. More recently, authors have focused on using metaanalytical methods for quantification of surrogacy. In this article, we extend existing procedures for analysis based on the accelerated failure time model to this setting. An advantage of this approach relative to proportional hazards model is that it allows for analysis in the semicompeting risks setting, where we model the region where the surrogate endpoint occurs before the true endpoint. Several estimation methods and attendant inferential procedures are presented. In addition, between- and within-trial methods for evaluating surrogacy are developed; a novel principal components procedure is developed for quantifying trial-level surrogacy. The methods are illustrated by application to data from several studies in colorectal cancer.     

Regularized estimation for the accelerated failure time model

Summary .  In the presence of high-dimensional predictors, it is challenging to develop reliable regression models that can be used to accurately predict future outcomes. Further complications arise when the outcome of interest is an event time, which is often not fully observed due to censoring. In this article, we develop robust prediction models for event time outcomes by regularizing the Gehan's estimator for the accelerated failure time (AFT) model ( Tsiatis, 1996 , Annals of Statistics 18, 305–328) with least absolute shrinkage and selection operator (LASSO) penalty. Unlike existing methods based on the inverse probability weighting and the Buckley and James estimator ( Buckley and James, 1979 , Biometrika 66, 429–436), the proposed approach does not require additional assumptions about the censoring and always yields a solution that is convergent. Furthermore, the proposed estimator leads to a stable regression model for prediction even if the AFT model fails to hold. To facilitate the adaptive selection of the tuning parameter, we detail an efficient numerical algorithm for obtaining the entire regularization path. The proposed procedures are applied to a breast cancer dataset to derive a reliable regression model for predicting patient survival based on a set of clinical prognostic factors and gene signatures. Finite sample performances of the procedures are evaluated through a simulation study.     

Generalized F accelerated failure time model for mapping survival trait loci

As the two most popular models in survival analysis, the accelerated failure time (AFT) model can more easily fit survival data than the Cox proportional hazards model (PHM). In this study, we develop a general parametric AFT model for identifying survival trait loci, in which the flexible generalized F distribution, including many commonly used distributions as special cases, is specified as the baseline survival distribution. EM algorithm for maximum likelihood estimation of model parameters is given. Simulations are conducted to validate the flexibility and the utility of the proposed mapping procedure. In analyzing survival time following hyperoxic acute lung injury (HALI) of mice in an F(2) mating population, the generalized F distribution performed best among the six competing survival distributions and detected four QTLs controlling differential HALI survival.     

Semiparametric frailty models for clustered failure time data

We consider frailty models with additive semiparametric covariate effects for clustered failure time data. We propose a doubly penalized partial likelihood (DPPL) procedure to estimate the nonparametric functions using smoothing splines. We show that the DPPL estimators could be obtained from fitting an augmented working frailty model with parametric covariate effects, whereas the nonparametric functions being estimated as linear combinations of fixed and random effects, and the smoothing parameters being estimated as extra variance components. This approach allows us to conveniently estimate all model components within a unified frailty model framework. We evaluate the finite sample performance of the proposed method via a simulation study, and apply the method to analyze data from a study of sexually transmitted infections (STI).     

Kernel based methods for accelerated failure time model with ultra-high dimensional data

Background  

Most genomic data have ultra-high dimensions with more than 10,000 genes (probes). Regularization methods with L ₁ and L _p penalty have been extensively studied in survival analysis with high-dimensional genomic data. However, when the sample size n ≪ m (the number of genes), directly identifying a small subset of genes from ultra-high (m > 10, 000) dimensional data is time-consuming and not computationally efficient. In current microarray analysis, what people really do is select a couple of thousands (or hundreds) of genes using univariate analysis or statistical tests, and then apply the LASSO-type penalty to further reduce the number of disease associated genes. This two-step procedure may introduce bias and inaccuracy and lead us to miss biologically important genes.     

Effect of covariate omission in Weibull accelerated failure time model: A caution

The accelerated failure time model is presented as an alternative to the proportional hazard model in the analysis of survival data. We investigate the effect of covariates omission in the case of applying a Weibull accelerated failure time model. In an uncensored setting, the asymptotic bias of the treatment effect is theoretically zero when important covariates are omitted; however, the asymptotic variance estimator of the treatment effect could be biased and then the size of the Wald test for the treatment effect is likely to exceed the nominal level. In some cases, the test size could be more than twice the nominal level. In a simulation study, in both censored and uncensored settings, Type I error for the test of the treatment effect was likely inflated when the prognostic covariates are omitted. This work remarks the careless use of the accelerated failure time model. We recommend the use of the robust sandwich variance estimator in order to avoid the inflation of the Type I error in the accelerated failure time model, although the robust variance is not commonly used in the survival data analyses.     

Statistical optimization of parametric accelerated failure time model for mapping survival trait loci

Most existing statistical methods for mapping quantitative trait loci (QTL) are not suitable for analyzing survival traits with a skewed distribution and censoring mechanism. As a result, researchers incorporate parametric and semi-parametric models of survival analysis into the framework of the interval mapping for QTL controlling survival traits. In survival analysis, accelerated failure time (AFT) model is considered as a de facto standard and fundamental model for data analysis. Based on AFT model, we propose a parametric approach for mapping survival traits using the EM algorithm to obtain the maximum likelihood estimates of the parameters. Also, with Bayesian information criterion (BIC) as a model selection criterion, an optimal mapping model is constructed by choosing specific error distributions with maximum likelihood and parsimonious parameters. Two real datasets were analyzed by our proposed method for illustration. The results show that among the five commonly used survival distributions, Weibull distribution is the optimal survival function for mapping of heading time in rice, while Log-logistic distribution is the optimal one for hyperoxic acute lung injury.     

Maximum likelihood estimation for interval-censored data using a Weibull-based accelerated failure time model.

The accelerated failure time regression model is most commonly used with right-censored survival data. This report studies the use of a Weibull-based accelerated failure time regression model when left- and interval-censored data are also observed. Two alternative methods of analysis are considered. First, the maximum likelihood estimates (MLEs) for the observed censoring pattern are computed. These are compared with estimates where midpoints are substituted for left- and interval-censored data (midpoint estimator, or MDE). Simulation studies indicate that for relatively large samples there are many instances when the MLE is superior to the MDE. For samples where the hazard rate is flat or nearly so, or where the percentage of interval-censored data is small, the MDE is adequate. An example using Framingham Heart Study data is discussed.     

Characterization of genomic imprinting effects and patterns with parametric accelerated failure time model

Genomic imprinting, a genetic phenomenon of non-equivalent allele expression that depends on parental origins, has been ubiquitously observed in nature. It does not only control the traits of growth and development but also may be responsible for survival traits. Based on the accelerated failure time model, we construct a general parametric model for mapping the imprinted QTL (iQTL). Within the framework of interval mapping, maximum likelihood estimation of iQTL parameters is implemented via EM algorithm. The imprinting patterns of the detected iQTL are statistically tested according to a series of null hypotheses. BIC model selection criterion is employed to choose an optimal baseline hazard function with maximum likelihood and parsimonious parameters. Simulations are used to validate the proposed mapping procedure. A published dataset from a mouse model system was used to illustrate the proposed framework. Results show that among the five commonly used survival distributions, Log-logistic distribution is the optimal baseline hazard function for mapping QTL of hyperoxic acute lung injury (HALI) survival; under the log-logistic distribution, four QTLs were identified, in which only one QTL was inherited in Mendelian fashion, whereas others were imprinted in different imprinting patterns.     

Identifiability assumptions for missing covariate data in failure time regression models

Methods in the literature for missing covariate data in survival models have relied on the missing at random (MAR) assumption to render regression parameters identifiable. MAR means that missingness can depend on the observed exit time, and whether or not that exit is a failure or a censoring event. By considering ways in which missingness of covariate X could depend on the true but possibly censored failure time T and the true censoring time C, we attempt to identify missingness mechanisms which would yield MAR data. We find that, under various reasonable assumptions about how missingness might depend on T and/or C, additional strong assumptions are needed to obtain MAR. We conclude that MAR is difficult to justify in practical applications. One exception arises when missingness is independent of T, and C is independent of the value of the missing X. As alternatives to MAR, we propose two new missingness assumptions. In one, the missingness depends on T but not on C; in the other, the situation is reversed. For each, we show that the failure time model is identifiable. When missingness is independent of T, we show that the naive complete record analysis will yield a consistent estimator of the failure time distribution. When missingness is independent of C, we develop a complete record likelihood function and a corresponding estimator for parametric failure time models. We propose analyses to evaluate the plausibility of either assumption in a particular data set, and illustrate the ideas using data from the literature on this problem.     

Statistical modelling and forecasting annual sugarcane production in India: Using various time series models

This paper proposes a comparison of various time series forecasting models to forecast annual data on sugarcane production over 63 years from 1960 to 2022. In this research, the Mean Forecast Model, the Naive Model, the Simple Exponential Smoothing Model, Holt's model, and the Autoregressive Integrated Moving Average time series models have all been used to make effective and accurate predictions for sugarcane. Different scale-dependent error forecasting techniques and residual analysis have been used to examine the forecasting accuracy of these time series models. SE of Residuals, Root Mean Squared Error (RMSE), Mean Absolute Error (MAE) and Akaike's Information Criterion (AIC) are used to analyse the forecast's accuracy. The best model has been selected based on the predictions with the lowest value, according to the three-performance metrics of RMSE, MAE, and AIC. The estimated sugarcane production shows an increasing trend for the next 10 years and is projected to be 37,763.38 million tonnes in the year 2032. Further, empirical results support the plan and execution of viable strategies to advance sugarcane production in India to fulfil the utilisation need of the increasing population and further improve food security.     

Alcohol consumption and mortality: modelling risks for men and women at different ages

ObjectiveTo estimate the relation between alcohol consumption and risk of death, the level of alcohol consumption at which risk is least, and how these vary with age and sex.DesignAnalysis using published systematic reviews and population data.SettingEngland and Wales in 1997.ResultsA direct dose-response relation exists between alcohol consumption and risk of death in women aged 16-54 and in men aged 16-34. At older ages the relation is U shaped. The level at which the risk is lowest increases with age, reaching 3 units a week in women aged over 65 and 8 units a week in men aged over 65. The level at which the risk is increased by 5% above this minimum is 8 units a week in women aged 16-24 and 5 units a week in men aged 16-24, increasing to 20 and 34 units a week in women and men aged over 65, respectively.ConclusionsSubstantially increased risks of all cause mortality can occur even in people drinking lower than recommended limits, and especially among younger people.

What is already known on this topic

Non-drinkers and heavy drinkers have higher all cause mortality rates than light drinkers—the U shaped curveThe precise shape and location of the U are likely to depend on age and sex, but this has not been quantified

What this study adds

The level of alcohol consumption that carries the lowest mortality ranges from 0 in men and women aged under 35 to 3 units a week in women aged over 65 and 8 units a week in men aged over 65The level of alcohol consumption that carries a 5% increase in mortality increases with age from 8 to 20 units a week in women and from 5 to 34 units a week in menOur calculations were for England and Wales in 1997: nadirs are likely to be lower in the future and in countries with less ischaemic heart disease